U.S. patent number 9,319,556 [Application Number 13/223,298] was granted by the patent office on 2016-04-19 for method and apparatus for authenticating printed documents that contains both dark and halftone text.
This patent grant is currently assigned to KONICA MINOLTA LABORATORY U.S.A., INC.. The grantee listed for this patent is Wei Ming, Songyang Yu, Xiaonong Zhan. Invention is credited to Wei Ming, Songyang Yu, Xiaonong Zhan.
United States Patent |
9,319,556 |
Zhan , et al. |
April 19, 2016 |
Method and apparatus for authenticating printed documents that
contains both dark and halftone text
Abstract
A document authentication method determines the authenticity of
a target hardcopy document, which purports to be a true copy of an
original hardcopy document. The method compares a binarized image
of the target document with a binarized image of the original
document which has been stored in a storage device. The image of
the original document is generated by binarizing a scanned
grayscale image of the original document. Halftone and non-halftone
text areas in the grayscale image area separated, and the two types
of text are separately binarized. The non-halftone text areas are
then down-sampled. During authenticating, a scanned grayscale image
of the target document is binarized by separating halftone and
non-halftone text areas and binarizing them separately, and then
down-sampling the non-halftone text areas. The binarized images of
the target document and the original document are compared to
determine the authenticity of the target document.
Inventors: |
Zhan; Xiaonong (Foster City,
CA), Ming; Wei (Cupertino, CA), Yu; Songyang (Foster
City, CA) |
Applicant: |
Name |
City |
State |
Country |
Type |
Zhan; Xiaonong
Ming; Wei
Yu; Songyang |
Foster City
Cupertino
Foster City |
CA
CA
CA |
US
US
US |
|
|
Assignee: |
KONICA MINOLTA LABORATORY U.S.A.,
INC. (San Mateo, CA)
|
Family
ID: |
47743333 |
Appl.
No.: |
13/223,298 |
Filed: |
August 31, 2011 |
Prior Publication Data
|
|
|
|
Document
Identifier |
Publication Date |
|
US 20130050764 A1 |
Feb 28, 2013 |
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
H04N
1/4446 (20130101); H04N 1/403 (20130101); H04N
1/32101 (20130101); H04N 2201/3235 (20130101); H04N
2201/3271 (20130101); H04N 2201/3269 (20130101); H04N
2201/3226 (20130101) |
Current International
Class: |
G06K
15/00 (20060101); H04N 1/32 (20060101); H04N
1/403 (20060101); H04N 1/44 (20060101); G06K
9/34 (20060101); G06K 9/48 (20060101); H04N
1/40 (20060101) |
References Cited
[Referenced By]
U.S. Patent Documents
Foreign Patent Documents
|
|
|
|
|
|
|
H06-231306 |
|
Aug 1994 |
|
JP |
|
2005-326971 |
|
Nov 2005 |
|
JP |
|
Other References
Japanese Office Action, dated May 20, 2014, in related Japanese
patent application, No. JP 2011-206636. cited by applicant .
OTSU, "A Threshold Selection Method from Gray-Level Histograms,"
IEEE Transactions on Systems, Man, and Cybernetics, vol. 9, No. 1,
1979, pp. 62-66. cited by applicant .
Rosenfield et al., "Thresholding using Relaxation", IEEE
Transactions on Pattern Analysis and Machine Intelligence, vol. 3,
No. 5, 1981, pp. 598-606. cited by applicant .
Shapiro et al., "An Adaptive Method for Image Thresholding",
Proceedings of the 11th IAPR Interational Conference on Pattern
Recognition, 1992, pp. 696-699. cited by applicant .
Niblack, "An introduction to Image Processing", Prentice-Hall,
Englewood Cliffs, 1986, pp. 115-116. cited by applicant .
Sauvola et al., "Adaptive document image binarization", Pattern
Recognition, vol. 33, 2000, pp. 225-236. cited by applicant .
Kim et al., "Document image binarization based on topographic
analysis using a water flow model", Pattern Recognition vol. 35
2002, pp. 265-277. cited by applicant .
Pratt, "Digital Image Processing", Fourth Edition, John Wiley &
Sons, Inc., 2007, pp. 624. cited by applicant .
Fisher et al., "A Rule-Based System for Document Image
Segmentation", Jun. 1990, Pattern Recognition Proceedings, 10th
International Conference, vol. 1, pp. 567-572. cited by applicant
.
USPTO Office Action, dated Sep. 20, 2013, in U.S. Appl. No.
12/946,801. cited by applicant .
USPTO Office Action, dated Apr. 11, 2014, in U.S. Appl. No.
12/946,801. cited by applicant.
|
Primary Examiner: Thomas; Ashish K
Assistant Examiner: Bataille; Frantz
Attorney, Agent or Firm: Chen Yoshimura LLP
Claims
What is claimed is:
1. A document authentication method implemented in a data
processing system, comprising: (a) obtaining an original grayscale
image representing an original document, the original grayscale
image including one or more halftone text areas or light text areas
and one or more non-halftone text areas or dark text areas, wherein
each halftone text area includes a plurality of halftone dots, each
halftone dot being formed by a plurality of pixels, each pixel
being a smallest unit of the original grayscale image and each
pixel having a grayscale pixel value; (b) separating the halftone
or light text areas from the non-halftone or dark text areas of the
original grayscale image; (c) separately binarizing the halftone or
light text areas and the non-halftone or dark text areas generated
by step (b); (d) down-sampling only the binarized non-halftone or
dark text areas generated by step (c) which correspond to the
non-halftone or dark text areas of the original grayscale image,
without down-sampling the binarized halftone or light text areas
which correspond to the halftone or light text areas of the same
original grayscale image, whereby a binarized original image is
generated; (e) obtaining a target grayscale image representing a
hardcopy target document, the target grayscale image including one
or more halftone text areas and one or more non-halftone text
areas, wherein each halftone text area includes a plurality of
halftone dots, each halftone dot being formed by a plurality of
pixels, each pixel being a smallest unit of the target grayscale
image and each pixel having a grayscale pixel value; (f) separating
the halftone text areas and the non-halftone text areas of the
target grayscale image; (g) separately binarizing the halftone text
areas and the non-halftone text areas generated by step (f); (h)
down-sampling only the binarized non-halftone text areas generated
by step (g) which correspond to the non-halftone text areas of the
target grayscale image, without down-sampling the binarized
halftone text areas which correspond to the halftone text areas of
the same target grayscale image, whereby a binarized target image
is generated; and (i) comparing the binarized target image with the
binarized original image to determine whether the target document
is an authentic copy of the original document.
2. The method of claim 1, wherein step (a) includes scanning the
hardcopy original document to generate the original grayscale
image, and step (e) includes scanning the hardcopy target document
to generate the target grayscale image.
3. The method of claim 1, further comprising, after step (a),
printing the original grayscale image to generate a copy of the
original document.
4. The method of claim 3, further comprising: after step (d),
storing the binarized original image in a storage device; and
before step (i), retrieving the stored binarized original image
from the storage device.
5. The method of claim 1, wherein step (b) includes: identifying
text characters in the original grayscale image; classifying each
identified text character as either a halftone text character or a
non-halftone text character based on a topological analysis of the
text character; and dividing the original grayscale image into
halftone text areas containing only halftone text characters and
non-halftone text areas containing non-halftone text characters;
where step (c) includes: binarizing each halftone text area using
pixel value statistics calculated from pixels in that area only;
wherein step (f) includes: identifying text characters in the
target grayscale image; classifying each identified text character
as either a halftone text character or a non-halftone text
character based on a topological analysis of the text character;
and dividing the target grayscale image into halftone text areas
containing only halftone text characters and non-halftone text
areas containing non-halftone text characters; and where step (g)
includes: binarizing each halftone text area using pixel value
statistics calculated from pixels in that area only.
6. The method of claim 1, wherein in steps (d) and (h), two or more
binarized non-halftone or dark text areas are down-sampled at
different down-sampling rates.
7. The method of claim 1, wherein step (i) includes comparing the
non-halftone text areas of the binarized target image and the
binarized original image and comparing the halftone text areas of
the binarized target image and the binarized original image using
different image comparison methods.
8. A computer program product comprising a computer usable
non-transitory medium having a computer readable program code
embedded therein for controlling a data processing apparatus, the
computer readable program code configured to cause the data
processing apparatus to execute a document authentication process
which comprises: (a) obtaining an original grayscale image
representing an original document, the original grayscale image
including one or more halftone text areas and one or more
non-halftone text areas, wherein each halftone text area includes a
plurality of halftone dots, each halftone dot being formed by a
plurality of pixels, each pixel being a smallest unit of the
original grayscale image and each pixel having a grayscale pixel
value; (b) separating the halftone text areas and the non-halftone
text areas of the original grayscale image; (c) separately
binarizing the halftone text areas and the non-halftone text areas
generated by step (b); and (d) down-sampling only the binarized
non-halftone or dark text areas generated by step (c) which
correspond to the non-halftone or dark text areas of the original
grayscale image, without down-sampling the binarized halftone or
light text areas which correspond to the halftone or light text
areas of the same original grayscale image, whereby a binarized
original image is generated.
9. The computer program product of claim 8, wherein the process
further comprises: (e) obtaining a target grayscale image
representing a hardcopy target document, the target grayscale image
including one or more halftone text areas and one or more
non-halftone text areas, wherein each halftone text area includes a
plurality of halftone dots, each halftone dot being formed by a
plurality of pixels, each pixel being a smallest unit of the target
grayscale image and each pixel having a grayscale pixel value; (f)
separating the halftone text areas and the non-halftone text areas
of the target grayscale image; (g) separately binarizing the
halftone text areas and the non-halftone text areas generated by
step (f); (h) down-sampling only the binarized non-halftone text
areas generated by step (g) which correspond to the non-halftone
text areas of the target grayscale image, without down-sampling the
binarized halftone text areas which correspond to the halftone text
areas of the same target grayscale image, whereby a binarized
target image is generated; and (i) comparing the binarized target
image with the binarized original image to determine whether the
target document is an authentic copy of the original document.
10. The computer program product of claim 9, wherein the process
further comprises: after step (d), storing the binarized original
image in a storage device; and before step (i), retrieving the
stored binarized original image from the storage device.
11. The computer program product of claim 9, wherein step (b)
includes: identifying text characters in the original grayscale
image; classifying each identified text character as either a
halftone text character or a non-halftone text character based on a
topological analysis of the text character; and dividing the
original grayscale image into halftone text areas containing only
halftone text characters and non-halftone text areas containing
non-halftone text characters; where step (c) includes: binarizing
each halftone text area using pixel value statistics calculated
from pixels in that area only; wherein step (f) includes:
identifying text characters in the target grayscale image;
classifying each identified text character as either a halftone
text character or a non-halftone text character based on a
topological analysis of the text character; and dividing the target
grayscale image into halftone text areas containing only halftone
text characters and non-halftone text areas containing non-halftone
text characters; and where step (g) includes: binarizing each
halftone text area using pixel value statistics calculated from
pixels in that area only.
12. The computer program product of claim 9, wherein in steps (d)
and (h), two or more binarized non-halftone or dark text areas are
down-sampled at different down-sampling rates.
13. The computer program product of claim 9, wherein step (i)
includes comparing the non-halftone text areas of the binarized
target image and the binarized original image and comparing the
halftone text areas of the binarized target image and the binarized
original image using different image comparison methods.
14. A computer program product comprising a computer usable
non-transitory medium having a computer readable program code
embedded therein for controlling a data processing apparatus, the
computer readable program code configured to cause the data
processing apparatus to execute a document authentication process
which comprises: (a) obtaining a target grayscale image
representing a hardcopy target document, the target grayscale image
including one or more halftone text areas and one or more
non-halftone text areas, wherein each halftone text area includes a
plurality of halftone dots, each halftone dot being formed by a
plurality of pixels, each pixel being a smallest unit of the
original grayscale image and each pixel having a grayscale pixel
value; (b) separating the halftone text areas and the non-halftone
text areas of the target grayscale image; (c) separately binarizing
the halftone text areas and the non-halftone text areas generated
by step (b); (d) down-sampling only the binarized non-halftone text
areas generated by step (c) which correspond to the non-halftone
text areas of the target grayscale image, without down-sampling the
binarized halftone text areas which correspond to the halftone text
areas of the same target grayscale image, whereby a binarized
target image is generated; and (e) comparing the binarized target
image with a binarized image of an original document to determine
whether the target document is an authentic copy of the original
document.
15. The computer program product of claim 14, wherein step (b)
includes: identifying text characters in the original grayscale
image; classifying each identified text character as either a
halftone text character or a non-halftone text character based on a
topological analysis of the text character; and dividing the
original grayscale image into halftone text areas containing only
halftone text characters and non-halftone text areas containing
non-halftone text characters; where step (c) includes: binarizing
each halftone text area using pixel value statistics calculated
from pixels in that area only.
16. The computer program product of claim 14, wherein step (e)
includes comparing the non-halftone text areas of the binarized
target image with corresponding areas of the binarized original
image and comparing the halftone text areas of the binarized target
image with corresponding areas of the binarized original image
using different image comparison methods.
Description
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates to a method for authenticating a copied
document by comparing a scanned image of the copied document with a
scanned image of the original hardcopy document. In particular, it
relates to a method for authenticating a copied document that
contains both dark and halftone text.
2. Description of Related Art
In situations where a hardcopy document (referred to as the
original document in this disclosure) is copied, e.g. by using a
photocopier, and the copy is distributed and circulated, there is
often a need to determine whether a purported true copy (referred
to as the target document in this disclosure) is authentic, i.e.,
whether the copied document has been altered while it was in
circulation. A goal in many document authentication methods is to
detect what the alterations (additions, deletions) are.
Alternatively, some document authentication methods determine
whether or not the document has been altered, without determining
what the alterations are.
Various types of document authentication methods are known. One
type of document authentication method performs a digital image
comparison of a scanned image of the target document with an image
of the original document. In such a method, a digital image of the
original document is stored in a storage device when the copy is
made. Later, the target document is scanned, and the stored image
of the original document is retrieved from the storage device to
compare with the image of the target document. In addition, certain
data representing or relating to the original document, such as a
document ID, is also stored in the storage device. The same data is
encoded in barcodes which are printed on the copied document when
the copy is made, and can be used to assist in document
authentication.
With the desire for high quality text and pictures, printed
documents are typically scanned at a high resolution, for example,
often at 600 dpi (dots per inch) or higher. Thus, document
authentication by image comparison can be computationally
intensive.
In a printed hardcopy document, gray or light color text or image
is often printed using a halftone method. For example, a printed
gray area will contain a plurality of black dots of ink or toner,
the sizes and/or density of the dots being dependent on the
gray-scale value of the gray area. Similarly, light colored areas
are printed by printing halftone patterns of color saturated dots.
In this disclosure, text printed by a halftone method is referred
to as halftone text. Image comparison for documents containing both
non-halftone (also referred to as dark or black) text and halftone
text may present additional challenge.
SUMMARY
The present invention is directed to an improved method for
authenticating a copied document that contains both non-halftone
text and halftone text.
An object of the present invention is to provide a document
authentication method with improved performance.
Additional features and advantages of the invention will be set
forth in the descriptions that follow and in part will be apparent
from the description, or may be learned by practice of the
invention. The objectives and other advantages of the invention
will be realized and attained by the structure particularly pointed
out in the written description and claims thereof as well as the
appended drawings.
To achieve these and/or other objects, as embodied and broadly
described, the present invention provides a document authentication
method implemented in a data processing system, which includes: (a)
obtaining an original grayscale image representing a hardcopy
original document; (b) separating halftone text areas and
non-halftone text areas in the original grayscale image; (c)
separately binarizing the halftone text areas and the non-halftone
text areas generated by step (b); (d) down-sampling the binarized
non-halftone text areas generated by step (c), whereby a binarized
original image is generated; (e) obtaining a target grayscale image
representing a hardcopy target document; (f) separating halftone
text areas and non-halftone text areas in the target grayscale
image; (g) separately binarizing the halftone text areas and the
non-halftone text areas generated by step (f); (h) down-sampling
the binarized non-halftone text areas generated by step (g),
whereby a binarized target image is generated; and (i) comparing
the binarized target image with the binarized original image to
determine whether the target document is an authentic copy of the
original document.
In another aspect, the present invention provides a computer
program product comprising a computer usable non-transitory medium
(e.g. memory or storage device) having a computer readable program
code embedded therein that causes a data processing apparatus to
perform the above method or parts thereof.
It is to be understood that both the foregoing general description
and the following detailed description are exemplary and
explanatory and are intended to provide further explanation of the
invention as claimed.
BRIEF DESCRIPTION OF THE DRAWINGS
FIGS. 1 and 2 schematically illustrate a document authentication
method according to an embodiment of the present invention. FIG. 1
illustrates a document registration stage and FIG. 2 illustrates an
authentication stage of the method.
FIGS. 3(a) and 3(b) illustrate exemplary grayscale images of
non-halftone text and halftone text.
FIGS. 3(c) and 3(d) illustrate exemplary binarized images
corresponding to the images in FIGS. 3(a) and 3(b).
FIG. 4 illustrates a method for separating halftone and
non-halftone text according to an embodiment of the present
invention.
FIG. 5 illustrates a system in which embodiments of the present
invention may be implemented.
DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS
Embodiments of the present invention provide a document
authentication method based on digital image comparison. In
particular, the method is suited for authenticating copied
documents that contains both non-halftone text and halftone
text.
FIG. 5 illustrates a system that may be used to implement the
document authentication method according to embodiments of the
present invention. The system includes one or more copiers 101,
scanners 102, printers 103, servers 104, mass storage devices 106,
etc. It may also include other components such as one or more
client computers 105, etc. The copiers, scanners, or printers may
be all-in-one devices, i.e., devices that combine a printing
section and scanning section in a single device and can perform
scanning, printing, and copying functions. Each of the copiers 101,
scanners 102, printers 103, servers 104, clients 105 etc. may
include a processor with associated memories which can carry out
data processing functions by executing programs stored in the
memory (these devices or a collection of them may be more generally
referred to as data processing apparatus or system). These
components are connected to each other by a network 107 and may be
located at distributed locations. The copier 101 may be used to
make a hardcopy of an original hardcopy document, the printer 103
may be used to print a hardcopy of an original electronic document,
and the scanner 102 or the copier 101 may be used to scan a copied
document (target document), as will be described later. Various
parts of the authentication method may be carried out by the server
104, the copier 101, the scanner 102, or the client 105, etc.
The authentication method according to embodiments of the present
invention includes a document registration stage and an
authentication stage. In the document registration stage, a copier
makes a hardcopy (i.e. on a physical medium such a paper) copy of
an original document. The original document may be a hardcopy form
or an electronic form. The copy may be made by scanning the
original hardcopy document and then printing a copy from the
scanned image, or printing the electronic original document. An
original document image, which is either the scanned image of the
hardcopy original document generated by the copier or an image
generated directly from the electronic original document, is
processed by a data processing apparatus and the resulting data is
stored in the storage device 106 (i.e. the original document is
registered in the system for later use). Details of the document
registration stage will be described with reference to FIG. 1.
Later, a user may submit a copied document (the target document)
for authentication by scanning the target document using a scanner
102 or copier 101, and causing a data processing apparatus to
retrieve the stored data from the storage device 106 and to perform
image comparison. Details of the authentication stage will be
described with reference to FIG. 2.
Referring to FIG. 1, in the document registration stage, an
original grayscale image representing the original document is
obtained (step S11). If the original document is in a hardcopy
form, step S11 involves scanning the hardcopy to generate an
original grayscale image; if the original document is in an
electronic form, step S11 involves generating a grayscale image
from the electronic original document if it is not already in an
image format. The original grayscale image is printed to generate a
copied document, i.e. a hard copy of the original document (step
S13).
Note that when the original document is a hardcopy containing
halftone text, then both the scanned grayscale image and the copied
hardcopy document will contain similar halftone text. If the
original document is in an electronic format and contains grey or
light colored text, the printed hardcopy document will contain
halftone text, while the original image (grayscale image directly
generated from the electronic original document) will contain gray
or light colored text with no halftone effect.
In addition, document management information, such as document ID,
is generated and encoded in barcode (step S12), which is also
printed on the copied document in step S13. The document ID will
aid in retrieval of the stored document during the authentication
stage. Optionally, other document management information may also
be encoded in the bar code, such as time of creation of the copy,
identity of the user who created the copy, etc., but this is not
critical because such information can be stored in the storage
device along with the image if desired.
The original grayscale image is processed in steps S14 to S17. More
specifically, in the case the original document was a hardcopy,
halftone text and non-halftone text in the grayscale image are
separated, and in the case the original document was electronic,
gray or light colored text (collectively referred to as light text
for convenience) is separated from black or dark text (collectively
referred to as dark text for convenience) (step S14). Different
types of text are binarized separately (step S15).
Separating light text from dark text is straightforward and can be
easily implemented by those skilled in the art. The description
below focuses on separating halftone text from non-halftone
text.
As is generally known, when a hardcopy document containing halftone
text is scanned at a sufficiently high resolution, in particular,
when the pixel size of the scan is substantially smaller than the
halftone dots, the halftone dots are typically visible in the
scanned image. Exemplary grayscale images of non-halftone and
halftone text are shown in FIGS. 3(a) and 3(b). FIG. 3(b) is an
enlarged view illustrating a scanned image of text when the text
was printed as a halftone gray color. In this illustration, the
halftone dots are clearly visible, and the scanned pixels in the
halftone text have various grayscale pixel values. Pixels located
within a halftone dot tend to have darker gray pixel values, and
pixels located in areas between adjacent halftone dots tend to have
lighter gray or white pixel values. The varying gray pixel values
arise from the limited sensitivity or accuracy of the scanner used
to scan the document. As a comparison, FIG. 3(a) is an enlarged
view illustrating a scanned image of text when the text in the
document was printed as non-halftone (black or dark) text. There
are no visible halftone dots. Further, the pixel values within the
text area are a black value or close to a black value. Although
black and gray text is used in these examples, the same effects
exist in color images, where the text may be printed as color
saturated text (similar to black text) or light colored text
(similar to light gray text, using a halftone method). The
descriptions below use black as an example, but the method
described herein is applicable to colored text as well.
When a printed document contains both non-halftone text and
halftone text, conventional binarization methods often lead to
unsatisfactory results for binarizing the halftone text. In
embodiments of the present invention, the document is separated
into non-halftone text and halftone text (step S14), and the two
types of text are binarized separately (step S15). Any suitable
methods may be used to perform these two steps; the method
described below is disclosed in commonly owned, co-pending U.S.
patent application Ser. No. 12/946,801, filed Nov. 15, 2010.
As shown in FIG. 4, this method separates halftone text from
non-halftone text based on an analysis of the topological features
of the text characters. First, the scanned document image is
initially binarized without regard to the presence of halftone text
(step S41). This step may use any suitable local or global textual
binarization method. Then, connected image components in the
initial binary image generated by step S41 are extracted (step
S42). In a binary image, a connected image component is a connected
group of pixels of the same pixel value (e.g. black). Methods for
extracting connected image components are generally known. Each
connected image component extracted in this step is potentially a
text character. Steps S41 and S42 collectively can be referred to
as a step of identifying text characters in the scanned document
image.
Then, each connected image component (text character) is classified
as either a halftone text character or a non-halftone text
character based on an analysis of its topology features (steps S43
and S44). Because of the halftone process, connected image
components for halftone text contain many more holes than connected
image components for non-halftone text. A hole is an area of white
pixels enclosed by black pixels of a connected image component. In
a preferred embodiment, this feature is used to classify each
connected image component (character). In a particular embodiment,
the Euler number is used as a criterion for classifying halftone
text characters and non-halftone text characters. The Euler number
(an integer) is a measure of the topology of an image. It is
defined as E=C-H, where E is the Euler number, C is the number of
connected image components in the image and H is the number of
holes. The Euler number for a text character (e.g. English letters
and Arabic numerals) is typically 1, 0 or -1. For example,
characters "C", "A" and "8" have Euler numbers of 1, 0 and -1,
respectively. Because binarized halftone text typically contains
many holes, the Euler number for a halftone text character
(including numerals) is generally much smaller than -1. For
example, the binarized halftone text character "N" shown in FIG.
4(a) has an Euler number of -43. As such, the Euler number is a
good measure for halftone text detection.
In step S43, the Euler number for each detected text character is
calculated. In step S44, each connected image component is
classified as either a halftone text character or a non-halftone
text character based on its Euler number. In one implementation, a
connected image component is classified as halftone text character
if it has an Euler number below a predefined value, such as -2.
Otherwise, it is classified as a non-halftone text character.
Based on the detected halftone text, the document image is divided
into non-halftone text areas which are areas containing only
non-halftone text and halftone text areas which are areas
containing only halftone text (step S45). This may be done, for
example, by applying binary morphological operations to the
non-halftone text and/or halftone text. Note that if the document
contains images or graphics, they are separately treated for
authentication which is not described in this disclosure.
Referring back to FIG. 1, after the non-halftone or dark text and
halftone or light text are separated in step S14, the non-halftone
or dark text and halftone or light text areas are binarized
separately (step S15). For example, pixel value statistics of the
non-halftone or dark text and halftone or light text areas can be
separately obtained and applied for binarization. Preferably, each
halftone or light text area is binarized using pixel value
statistics calculated from pixels in that area only. Exemplary
images of binarization result corresponding to the images in FIGS.
3(a) and 3(b) are shown in FIGS. 3(c) and 3(d).
Then, for non-halftone or dark text areas, the binarized images are
down-sampled, for example, by two times or three times (step S16).
In some embodiments, the down-sampling rate may be dynamically
determined for different areas of the document. For example, areas
of larger sized text, such as titles, headlines, etc., may be
down-sampled at a higher rate, such as six times.
In a preferred embodiment, only non-halftone or dark text areas are
down-sampled. In other words, the resolution of the halftone or
light text areas is maintained. This is because the binarized
images of the non-halftone text areas typically have higher quality
than the halftone areas, and down-sampling would not adversely
affect the quality of the resulting binary images and subsequent
authentication performance. On the other hand, as seen by comparing
FIGS. 3(c) and 3(d), the binarized images of halftone text (FIG.
3(d)) tend to be less smooth than the binarized images of
non-halftone text (FIG. 3(c)), and down-sampling halftone text
areas to the same degree as the non-halftone text areas may
adversely affect the image quality and subsequent authentication
performance. In other embodiments, the halftone or light text areas
are also down-sampled, but by a lesser amount than the non-halftone
or dark text areas. This may be suitable when the scanned image has
a sufficiently high resolution so that even after down-sampling,
the binarized halftone text can maintain sufficient quality.
The binarized image of the original document generated as a result
of steps S15 and S16 is stored in a storage device (step S17). The
image is stored in association with the document management
information, such as the document ID, to facilitate image retrieval
during the authentication stage.
Note that after the non-halftone or dark text areas are
down-sampled, different areas of the binarized image will have
different resolutions and the image is no longer a simple M.times.N
binary image. Thus, when storing the binarized image, the various
areas may be stored individually as binary images, along with
information about their positions.
In step S17, information regarding the down-sampling rates used to
down-sample the non-halftone or dark text area (and halftone or
light text areas if applicable) is also stored in the storage
device along with the binarized image. Other optional information
that may be stored in the storage device includes the resolution at
which the original document was scanned. Alternatively or in
addition, the information mentioned above may be encoded in the
barcode which is printed on the copied document.
Optionally, the grayscale image of the scanned original document
generated in step S11 can also be stored in the storage device so
that the document may be re-printed when desired.
Note that the order of performance of steps S12 and S13 relative to
steps S14 to S17 is generally not important. Of course, if the
halftone text positions are encoded in the barcode, then steps S12
and S13 should be performed after step S14.
In the document registration stage, steps S14 to S17 may be
performed by the copier or printer, in which case the copier can
transmit the binarized image to the server or store it directly in
the storage device; or they may be performed by the server, in
which case the copier will transmit the grayscale image to the
server. Step S12 likewise may be performed by either the
copier/printer or the server. More generally, the data processing
steps S12 and S14 to S17 may be performed in a distributed manner
by several devices.
The authentication stage is described with reference to FIG. 2. The
target document is scanned to generate a target grayscale image
(step S21). The barcode contained in the target image is extracted
and decoded to obtain the information contained therein, including
the document ID (step S22). The document ID is then used to
retrieve the stored binarized image of the original document having
the same document ID from the storage device (step S23). Other
relevant information stored in the storage device, such as
positions of the halftone text, may also be retrieved in this
step.
The target grayscale image is processed to separate non-halftone
text and halftone text (step S24), and the non-halftone text and
halftone text are binarized separately (step S25). Steps S24 and
S25 may use generally the same method as steps S14 and S15 in the
document registration stage, except that (optionally) if the
positions of the halftone text in the original document is
available from the storage device or the barcode, they can be used
as an initial estimate for the positions of the halftone text in
the target grayscale image.
Then, for non-halftone text areas, as well as halftone text areas
if appropriate, the binarized images are down-sampled (step S26).
The down-sampling rates used for this step are the same as those
used to down-sample the binarized image of the original document,
which have been retrieved from the storage device and/or decoded
from the barcode.
The target binarized image generated by steps S25 and S26 is then
compared with the original binarized image retrieved from the
storage device to determine if the target document is authentic
(i.e. whether it has been altered). Each pair of corresponding text
areas in the two images are compared. Of course, if any text area
in one image does not have a corresponding text area in the other
image, it can be concluded that the target document is not
authentic.
The same or different comparison methods may be used to compare the
non-halftone text areas of the target image with the original image
and to compare the halftone text areas of the target image with the
original image. Any suitable image comparison methods may be used.
One well-known method for comparing similarity of two images uses
normalized cross-correlation. This method is applicable for both
halftone and non-halftone text areas of the target image.
Another image comparison method, described in commonly owned U.S.
Pat. No. 8,000,528, issued Aug. 16, 2011, involves segmenting the
original and target documents into paragraph, line, word and
character units, and comparing the two images at progressively
lower levers. The paragraph level comparison determines whether the
target and original images have the same number of paragraphs and
whether the paragraphs have the same sizes and locations; the line
level comparison determines if the target and original images have
the same number of lines and whether the lines have the same sizes
and locations; etc. This method may be applicable for both halftone
and non-halftone text areas.
Yet another image comparison method, described in commonly owned
U.S. Pat. No. 7,965,894, issued Jun. 21, 2011, involves a two-step
comparison. In the first step, the original and target images are
divided into connected image components and their centroids are
obtained, and the centroids of the image components in the original
and target images are compared. Each centroid in the target image
that is not in the original image is deemed to represent an
addition, and each centroid in the original image that is not in
the target image is deemed to represent a deletion. In the second
step, sub-images containing the image components corresponding to
each pair of matching centroids in the original and target images
are compared to detect any alterations. This method may be
applicable for both halftone and non-halftone text areas.
Yet another image comparison method, described in commonly owned,
co-pending U.S. patent application Ser. No. 13/053,618, filed Mar.
22, 2011, involves comparing pairs of text characters by analyzing
and comparing their shape features such as their Euler numbers,
aspect ratios of their bounding boxes, pixel densities, the
Hausdorff distance between the two characters, etc. For halftone
text areas in the target image (even if the corresponding area in
the original image contains only light text without halftone
effect), this method is likely to be inadequate. Even when the
halftone text has been separately binarized, characters may still
contain extra holes, so the Euler number of a character may be an
inaccurate representation of the character's shape features.
In the authentication stage, steps S24 to S27 may be performed by
the scanner, in which case the scanner can request the binarized
original image from the server or retrieve it directly from the
storage device; or they may be performed by the server, in which
case the scanner will transmit the target grayscale image to the
server. Step S22 likewise may be performed by either the scanner or
the server. More generally, the data processing steps S22 to S23
and S24 to S27 may be performed in a distributed manner by several
devices.
Note that the order of performance of steps S22 to S23 relative to
steps S24 to S26 is generally not important. Of course, if any
stored information is used to aid in binarization, then steps S22
and S23 should be performed before step S24 to S26.
Although not shown in the drawings, various post-processing steps
may be carried out, such as generating a difference map between the
original image and the target image if any alteration is detected,
displaying the detection result to the user, etc. These steps may
be implemented using any suitable methods.
An advantage of the document authentication method according to
embodiments of the present invention is that, by separating
halftone or light and non-halftone or dark text, the two types of
text may be treated differently during the document registration
stage and authentication stage, so that the non-halftone or dark
text may down-sampled to a greater degree than the halftone or
light text. This improves speed of the authentication stage without
sacrificing reliability.
In the methods shown in FIGS. 1 and 2, the down-sampling of
non-halftone or dark areas of the original image is performed
during the document registration stage and the resulting image is
stored in the storage device. Alternatively, the down-sampling of
the original image (step S16) is not performed in the document
registration stage, but is performed in the authentication stage.
In this alternative approach, the binarized image generated by step
S15 is stored in the storage device, along with information
regarding the positions of the halftone or light text areas. Then,
during the authentication stage, the stored binarized original
image retrieved from the storage device, along with the position
information, and the image is down-sampled before comparing to the
binarized and down-sampled target image. Because down-sampling of
the non-halftone or dark areas is not computationally intensive,
this alternative embodiment does not significantly impact the
performance of the authentication stage.
In the methods shown in FIGS. 1 and 2, the binarization steps of
the original image (steps S14 to S16 of FIG. 1) are performed in
the document registration stage and the binarized original image is
stored in a storage device. An advantage of this approach is that
binarization is performed only once for each original document, and
the result can be used multiple times later if needed. Another
advantage is that the speed of the authentication stage is faster.
Alternatively, the original grayscale image may be stored in the
storage device without binarizing it, and later when a request for
authentication is received, the original grayscale image is
binarized and then compared to the binarized target image.
It will be apparent to those skilled in the art that various
modification and variations can be made in the document
authentication method of the present invention without departing
from the spirit or scope of the invention. Thus, it is intended
that the present invention cover modifications and variations that
come within the scope of the appended claims and their
equivalents.
* * * * *